The basic idea in this patch is to create an MediaDecoderOwner
interface which nsHTMLMediaElement would implement, and put everything
needed by nsMediaDeocder on that interface. In addition to that,
there are a number of other cleanup patches which enables us to
eliminate many of the nsHTMLMediaElement.h #includes in the media code.
This patch moves the various Is{Format}Enabled methods to
nsMediaDecoder. These methods are all implemented on top of prefs, and
they logically belong to the decoder class, because it needs to know
what media types it should be able to decode. This also helps reduce
the dependency on nsHTMLMediaElement.h.
This patch simply moves the MetadataTags typedef to nsMediaDecoder.h to
reduce the dependencies on nsHTMLMediaElements.h. The rest of the
changes are made to make this compile.
The code to handle MediaStream URIs is almost the same as for Blobs, so share it.
nsHostObjectProtocolHandler is modified a bit to simplify method names. Also
we make nsHostObjectProtocolHandler::AddDataEntry take responsibility for
generating the URI to avoid duplicating that code later.
We need separate subclasses for each kind of host object protocol handler,
but we don't need separate subclasses for each kind of host object URI.
--HG--
rename : content/base/public/nsBlobProtocolHandler.h => content/base/public/nsHostObjectProtocolHandler.h
rename : content/base/src/nsBlobProtocolHandler.cpp => content/base/src/nsHostObjectProtocolHandler.cpp
rename : content/base/src/nsBlobURI.cpp => content/base/src/nsHostObjectURI.cpp
rename : content/base/src/nsBlobURI.h => content/base/src/nsHostObjectURI.h
extra : rebase_source : bfb1d7ea2813100a35f9a2054a67584e38f5f330
This patch implements most of the semantics of AudioParam in a
templatized base class called AudioEventTimeline. This is done in order
to make it possible to test this code from C++ without the need of
linking to libxul. Basically we take everything that could depend on
libxul or is not suitable for a genertic implementation and put it in a
traits like type as a template argument. Then the test creates mock
objects that conform to the argument interfaces, and tests the logic of
AudioEventTimeline.
Also fixes what I think is a bug in InterleaveAndConvertBuffer converting S16 to S16.
Instead of clamping the volume, we should handle arbitrary volumes by falling back
to the float conversion path.
Replace nsAudioStream::Format with an AUDIO_OUTPUT_FORMAT enum value so we
can use it as a template parameter.
Introduce AudioSampleTraits<AudioSampleFormat> to give us access to the C++ type
corresponding to an enum value.
Move SampleToFloat/FloatToSample to AudioSampleFormat.h.
Introduce ConvertAudioSamples and ConvertAudioSamplesWithScale functions
and use them from various places.
Moves AudioDataValue to AudioSampleFormat.h. The name isn't great, but it'll do.
